20 likes | 396 Views
MIMO-OFDM with ESPAR Antenna. Diego Javier Reinoso Chisaguano , Takeshi Higashino and Minoru Okada. Nara Institute of Science and Technology, Graduate School of Information Science Network Systems Laboratory. MIMO-OFDM with ESPAR antenna.
E N D
MIMO-OFDM with ESPAR Antenna Diego Javier ReinosoChisaguano, Takeshi Higashino and Minoru Okada • Nara Institute of Science and Technology, Graduate School of Information Science • Network Systems Laboratory MIMO-OFDM with ESPAR antenna ESPAR antennawith periodically changing directivity • In this scheme the directivity of the ESPAR antenna is changed by an oscillator which frequency is the OFDM symbol rate. • The periodic variation of the directivity causes Inter ICI in the received signal. • In the figure, (a) is the frequency non-shifted component, the positive and negative frequency shifted components are shown in (b), (c), and the total received signal in (d) • ESPAR(Electronically Steerable Passive Array Radiator). • It is a small size and low power consumption antenna. • It is composed by a radiator element connected to the RF front-end and one or more parasitic(passive) elements terminated by variable capacitances. System model of MIMO-OFDM with ESPAR antenna • Compared to the conventional MIMO-OFDM 2x2 systems, MIMO-OFDM with ESPAR antenna gives additional diversity gain and improves the bit error rate performance without increasing the number of RF front-end circuits. • a) Transmitter • b) Receiver • For every receiver it uses a 2-elements ESPAR antenna with periodically changing directivity • It uses the low complexity MMSE sparse-SQRD algorithm for the detection process. • Based in the WLAN IEEE 802.11n standard Compressed Sensing based channel estimation • For MIMO-OFDM with ESPAR antenna, when the pilot symbol is transmitted, the vector of received symbols at the i-th receiver and after the FFT block is given by • Compressed Sensing (CS) is a set of new algorithms that allows the reconstruction of sparse signals from much fewer measurements. • To exploit the sparsity of the channel impulse response, the previous equation can be expressed as ui = [G-1P1 , G0P1 , G1P1 , G-1P2 , G0P2 , G1P2] r z + = • Channel impulse response F :N x L matrix part of the Fourier matrix. N : numberof sub-carriers L : number of channel taps. r: observation vector of size n : n x p measurement matrix. (n<<p) : sparse unknown vector of size p z: noise vector • To solve the previous equation, Dantzig Selector (DS) or Orthogonal Matching Pursuit (OMP) algorithms can be used. Simulation Results Simulation Settings 16dB • For a BER of 10-3 and using perfect CSI, MIMO-OFDM with ESPAR antenna achieves an additional diversity gain of 16dB compared to a common MIMO 2x2 VBLAST system. • Using CS-based channel estimation with OMP we obtain better estimation accuracy and it improves the BER compared to the MMSE channel estimator.